Paul Gearon wrote:
Hi Pat,
On Thu, Jul 1, 2010 at 9:52 PM, Pat Hayes <[email protected]> wrote:
Hey, guys. It is perfectly fine to use OWL properties in RDF. The RDF specs
actually encourage this kind of semantic borrowing, it was always part of
the RDF design to have this happen. So no need to have a version of
owl:sameAs in the RDFS namespace. Just use the OWL one.
fwiw, I was thinking more from a developer standpoint, hitting one
simple spec doc and finding most of what they need to get modelling
without being faced with OWL FULL and DL and the vast amounts of docs +
reading. This is the main reason I personally mentioned :)
Yes, I know that borrowing terms is allowed. Indeed, it gets used every day.
The thing is that we're talking about maybe cleaning RDF up a little.
(emphasis on the "maybe" - though that's starting to look more
likely). In this case, it makes sense to me that a term for equality
would make it's way into RDFS, simply because there are a lot of use
cases where people are sticking to just that namespace, with the
single exception of owl:sameAs. Also from an aesthetics point of view,
equality is such a common concept that I'm surprised it wasn't already
lower in the stack.
Nothing in RDF *needs* to be changed. But if it does get updated, then
I think that it would be nice to clean things a little while all the
new features get added (such as named graphs).
Regards,
Paul Gearon
